GH-127456: pathlib ABCs: remove Parser.sep

[barneygale]

Remove Parser.sep, which is the protocol's only non-method member. Instead, we compute the separator from PathBase.__init_subclass__() by calling parser.join('a', 'a').strip('a'), and store the result on the subclass as cls._sep. We also set cls._case_sensitive according to parser.normcase('Aa') == 'Aa', which was previously done in a caching function.

• Issue: pathlib ABCs: add protocols for supporting types #127456

[encukou]

I see there's currently no distinction between pathlib-internal attribute names and names that are free for third-party subclasses to use.
Should there be a stronger namespace for these, like _pathlib_sep, _pathlib_globber & _pathlib_stack?
Or should _sep & co. be documented so users know to not override them?

[barneygale]

I suppose another option is to make them public as PurePath.separator and PurePath.is_case_sensitive

[encukou]

Hmm. What's the reason for removing sep from Parser, again? That's really the correct place for it.

Should there be a stronger namespace for these, like _pathlib_sep, _pathlib_globber & _pathlib_stack?

I guess that might be for another PR, since it affects _globber and methods as well.

I suppose another option is to make them public as PurePath.separator and PurePath.is_case_sensitive

Right, but then __init_subclass__ should only set them if they're not in __dict__.
It's probably best to keep them as private attributes. Or continue using @functools.cache to avoid magic leaking into the API -- e.g. here, you can't do:

classMyPath(PurePathBase): passMyPath.parser=MyParser()

[barneygale]

All good points. Really my only motivation was to make the Parser interface method-only - partly so that issubclass() checks work, and partly for likely-misguided aesthetic reasons. You've convinced me it isn't a good idea, thank you for your feedback as ever!